The aim of this project is to examine the role of extended family networks and intergenerational migration patterns in risk sharing. Two persistent trends the world has seen in the past two centuries are increasing market integration and greater movement of people across regions. In the absence of large, extended family databases, it has not been possible to assess the significance of interfamily responses to economic shocks and how intergenerational family structure changes in response to increases in market development. The proposed project seeks to examine these relationships through analysis of the largest intergenerational data set to have been compiled from historical Chinese genealogical sources to date. The source contains demographic and economic information on every generation of 41 separate lineages from Tongcheng County in Anhwei Province between the years 1500-1850. The 100,000 individuals in the dataset are representative of a random socio-economic sample of those who resided in Tongcheng, and comprise approximately 10 to 35 percent of the total population that lived in the county over the 350 years covered. The analysis will estimate how social and health variables--age-specific mortality rates, fertility, and age-of-marriage--are affected by risk-sharing behavior within the extended family, and it will model the extent to which insurance motives for migration determine family structure and formation. The project will also identify whether increasing market development over time tends to weaken these aspects of family linkages. In addition to conducting these studies, a principle long-term objective of the project is to lay the groundwork for establishing an expanded data set based on genealogies and intergenerational datasets from other regions in China or elsewhere. The collection and systematic tabulation of an intergenerational data set on a larger geographical scale will allow us to explore long-distance population movements, resource use, business practice, and public policy impacts of migration more generally. The dissemination of the results of this study will generate greater interest in research in this area and provide a foundation for the larger, more geographically inclusive, studies.